Cryptographic computations on general purpose graphics processing units

Abstract

One embodiment of the invention sets forth a computing system for performing cryptographic computations. The computing system comprises a central processing unit, a graphics processing unit, and a driver. The central processing requests a cryptographic computation. In response, the driver downloads microcode to perform the cryptographic computation to the graphics processing unit and the graphics processing unit executes microcode. This offloads cryptographic computations from the CPU. As a result, cryptographic computations are performed faster and more efficiently on the GPU, freeing resources on the CPU for other tasks.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS[0001]This application is a divisional of U.S. patent application Ser. No. 11 / 454,765, filed Jun. 15, 2006, which is hereby incorporated herein by reference.BACKGROUND OF THE INVENTION[0002]1. Field of the Invention[0003]Embodiments of the present invention relate generally to cryptography and general-purpose computing on graphics processing units and more specifically to cryptographic computations on graphics processing units.[0004]2. Description of the Related Art[0005]Modern Graphics processing units (GPUs) use millions of transistors to perform calculations related to 3D computer graphics. GPUs were initially used to accelerate the memory-intensive work of texture mapping and rendering polygons, later adding units to accelerate geometric calculations such as translating vertices into different coordinate systems. Recent developments in GPUs include support for programmable shaders which can manipulate vertices and textures with many of the ...

AI Technical Summary

Benefits of technology

The invention is a system and method for performing cryptographic computations using a CPU, GPU, and driver. The system can download microcode to the GPU and execute it, performing both integer and logical operations. The driver can divide requests into batches based on CPU speed and thread execution. The GPU can have multiple thread groups, each encrypting different parts of a input state simultaneously. The system also includes a pre-computed table to replace certain steps in the AES algorithm. The invention allows for faster and more efficient cryptographic computations on the GPU, freeing up resources on the CPU for other tasks.

Problems solved by technology

The lack of integers and associated operations, such as bit-shifts and bitwise logical operations (e.g., AND, OR, XOR, NOT) makes GPUs ill suited for many mathematically complex tasks, such as cryptography.

GPGPU computing presents additional challenges even for problems that map well to the GPU, because oftentimes GPU programming typically requires recasting the relevant computations into graphics terms.

Thus, harnessing the power of a GPU for general-purpose computation often requires a concerted effort by experts in both computer graphics and in the particular scientific or engineering domain.

The huge throughput required to perform cryptographic computations for AES will likely overwhelm most CPUs (e.g., bidirectional throughput of over 2 Gbits / s).

Images